756 



REPORT — 1897. 



•with many where they were not ; No. 10 contains a few tests of wroiight-iron 

 pipes used as cokimns. 



In the tests made at Watertown Arsenal we have a long series on wrought- 

 iron built up bridge columns. 



Timber Columns. 



1. ' Exec. Doc. 1,' 47th Congress, 1st Session, House. 



2. Report of tests on full-size wooden mill columns, by G. Lanza, 1882. 



3. 'Exec. Doc. 1,' 47th Congress, 2nd Session, Senate. 



4. ' Journal Assoc. Engineering Societies,' Nov. 1889. 



5. ' Technology Quarterly,' vol. viii., 1895. 



6. Bauschinger, ' Mittheilungen aus dem Konigl. Mech. Tech. Lab., Heft 9 

 and Heft 16. 



7. * Transactions Canadian Soc. Civil Engineers,' vol. ix., 1895. 



The tests cited in Nos. 1, 2, .3, and 4 were made at Watertown Arsenal, 

 and comprise a very extensive series of tests of full-size timber columns ; those 

 cited in No. 5 were made at the Massachusetts Institute of Technology, those in 

 No. 6 by Professor Bauschinger at Munich, and those in No. 7 by Professor 

 Bovey at McGill College, Montreal. 



In order to represent to the eye the results of these tests, and therefore to 

 enable us to discuss them, the following diagrams are presented : — 



1. A diagram showing the results of the tests of cast-iron mill columns. 



2. A series of four diagrams showing the results of the tests of wrought-iron 

 bridge columns, and also empirical formulae representing in each case the right- 

 hand portion of the curve, which is concave upwards. 



3. A series of four diagrams showing the results of the tests of timber 

 columns cited in Nos. 1, 2, 3, and 5. 



In 1 and 2 the abscissaj represent the ratio of length to least radius of 

 gyration, and in 3 the ratio of length to least diameter, while the ordinates 

 represent, in all the diagrams, the breaking loads per square inch of sectional 

 area. 



A study of these diagrams, and of the details of the tests which they represent, 

 gives us the facts in regard to the strength of full-size columns, and shows that 

 neither the experiments of Eaton Hodgkinson upon small samples nor the usual 



Cast-iron Columns from Pacific Mills. 



£0000 



60 65 70 75 80 85 30 35 /OO /OS //O //S /SO X?5 ./30 /35 /40 /45 /SO 



eoooo 



Abscissce, length divided by radius of gyration of smallest section. 

 Ordinates, breaking strengths per square inch of smallest section. 



■Riilpv or Gordon theories (so commonly quoted in the handbooks) are borne out by 

 the facts 



A perusal of all the diagrams show that, whenever the load on a column is so 

 applied that its resultant acts along the axis of the column, the breaking load per 

 square inch of sectional area is practically constant up to a certain ratio of lengtli 

 to radius of gyration, which in wrought-iron bridge columns varies from sixty to 

 eighty, and in a corresponding way to timber columns. 



